ClusterR (version 1.3.3)

predict_MBatchKMeans: Prediction function for Mini-Batch-k-means

Description

Prediction function for Mini-Batch-k-means

Usage

predict_MBatchKMeans(data, CENTROIDS, fuzzy = FALSE, updated_output = FALSE)
# S3 method for MBatchKMeans
predict(object, newdata, fuzzy = FALSE, ...)

Value

if fuzzy = TRUE the function returns a list with two attributes: a vector with the clusters and a matrix with cluster probabilities. Otherwise, it returns a vector with the clusters.

Arguments

data

matrix or data frame

CENTROIDS

a matrix of initial cluster centroids. The rows of the CENTROIDS matrix should be equal to the number of clusters and the columns should equal the columns of the data.

fuzzy

either TRUE or FALSE. If TRUE then prediction probabilities will be calculated using the distance between observations and centroids.

updated_output

either TRUE or FALSE. If TRUE then the 'predict_MBatchKMeans' function will follow the same output object behaviour as the 'predict_KMeans' function (if fuzzy is TRUE it will return probabilities otherwise it will return the hard clusters). This parameter will be removed in version 1.4.0 because this will become the default output format.

object, newdata, ...

arguments for the `predict` generic

Author

Lampros Mouselimis

Details

This function takes the data and the output centroids and returns the clusters.

Examples

data(dietary_survey_IBS)

dat = dietary_survey_IBS[, -ncol(dietary_survey_IBS)]

dat = center_scale(dat)

MbatchKm = MiniBatchKmeans(dat, clusters = 2, batch_size = 20, num_init = 5, early_stop_iter = 10)

pr = predict_MBatchKMeans(dat, MbatchKm$centroids, fuzzy = FALSE)